Related Experiment Video
Updated: May 11, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
2.4K
An automatic sustained attention prediction (ASAP) method for infants and toddlers using wearable device signals
Yisi Zhang1, A Priscilla Martinez-Cedillo2,3, Harry T Mason4,5
1Department of Psychological and Cognitive Sciences, Tsinghua University, Beijing, 100084, People's Republic of China.
Scientific Reports
|April 17, 2025
Summary
This study introduces an automatic sustained attention prediction (ASAP) method using wearable sensors to measure infant attention in natural settings. ASAP effectively detects sustained attention (SA) during free play, opening new avenues for developmental research.
Area of Science:
- Developmental Psychology
- Cognitive Neuroscience
- Wearable Technology
Background:
- Sustained attention (SA) is crucial for infant development, but lab-based research limits real-world applicability.
- Infant wearable technology offers a way to collect multimodal data, including physiological signals, in natural environments.
- Existing methods for measuring SA in infants often rely on controlled lab settings.
Purpose of the Study:
- To introduce and validate an automatic sustained attention prediction (ASAP) method using electrocardiogram (ECG) and accelerometer (Acc) signals.
- To assess the performance of ASAP in detecting SA in infants aged 6-36 months during naturalistic activities.
- To explore the relationship between infant SA and visual perceptual features (saliency, clutter) in real-world settings.
Main Methods:
- Developed the automatic sustained attention prediction (ASAP) method using ECG and accelerometer data from 75 infants (6-36 months).
- Trained ASAP on temporal and spectral features extracted from physiological signals.
- Validated ASAP's performance against human-coded SA periods derived from ECG and fixation data.
- Collected data during various activities, including free play, emulating natural environments.
Main Results:
- ASAP demonstrated consistent performance in detecting sustained attention (SA) across different infant age groups.
- ASAP's detection of SA showed no significant differences compared to human-coded SA periods.
- Infant visual saliency increased during attention periods, and this effect decreased with age.
- No age-related differences were found for inattention periods.
Conclusions:
- The ASAP method effectively detects sustained attention (SA) in infants during naturalistic free play using wearable sensor data.
- ASAP provides a reliable, non-invasive tool for studying infant cognitive development in real-world contexts.
- This technology enables unprecedented opportunities for understanding infant attention and its relationship with environmental factors.
